Auss Abbood

PhD student in Computation, Cognition and Language at the University of Cambridge.

profile_pic.jpg

Graduate Matriculation 2024

Lucy Cavendish College

I am Auss, a first-gen immigrant from Iraq, a cognitive scientist by training, sports enthusiast, and bassist. Since October 2024, I have been a PhD student with Prof Nigel Collier at the Language Technology Lab. There, I explore temporal reasoning in large language models (LLMs). I am particularly interested in utilising metacognition in LLMs to better understand temporal reasoning, identify failure modes, and improve performance on tasks that require a mix of skills.

Before my PhD, I have been working as a data scientist at the Robert Koch Institute informing political and medical stakeholders. There, I have been working on early detection of infectious diseases using natural language processing and time series analysis. I was in the founding team that developed Germany’s first surveillance system for intensive care unit capacities, the DIVI-Intensivregister. Later, I was responsible to track international outbreak situations, using epidemic modelling and data analyses. Outside of Germany, I worked on capacity building in digital epidemiology in Southern Africa and Central Asia.

News

Nov 11, 2026 I wrote a blog post reflecting on my time at the UK AI Security Institute and why AI evaluation is important but tricky to get right.
Nov 10, 2026 I co-facilitated a pre-congress workshop on AI and big data in infectious diseases at ICID 2026 in Madrid!
Oct 30, 2026 Check out the recent AI Evaluation Digest! It has been my favourite Substack and I now got the chance to contribute to it.
Oct 22, 2026 What does it take to reason about time? 🧮 Working with dates and clocks, or 🧩 understanding event order and cause and effect? Our new preprint tests that split through cross-benchmark transfer experiments and cognitive-demand analysis.
Oct 19, 2026 New side quest unlocked 🎧 Started a DJ Society at Lucy Cavendish College as Co-President. Gig announcements coming soon 🔥🎛️